﻿<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:trackback="http://madskills.com/public/xml/rss/module/trackback/" xmlns:wfw="http://wellformedweb.org/CommentAPI/" xmlns:slash="http://purl.org/rss/1.0/modules/slash/"><channel><title>BlogJava-专注创造价值-随笔分类-聚类算法研究</title><link>http://www.blogjava.net/fullfocus/category/31312.html</link><description /><language>zh-cn</language><lastBuildDate>Fri, 09 May 2008 21:18:46 GMT</lastBuildDate><pubDate>Fri, 09 May 2008 21:18:46 GMT</pubDate><ttl>60</ttl><item><title>focus聚类研究系列一-----熟悉现有项目基础（站在巨人的肩膀上）</title><link>http://www.blogjava.net/fullfocus/archive/2008/05/09/199608.html</link><dc:creator>fullfocus</dc:creator><author>fullfocus</author><pubDate>Fri, 09 May 2008 14:22:00 GMT</pubDate><guid>http://www.blogjava.net/fullfocus/archive/2008/05/09/199608.html</guid><wfw:comment>http://www.blogjava.net/fullfocus/comments/199608.html</wfw:comment><comments>http://www.blogjava.net/fullfocus/archive/2008/05/09/199608.html#Feedback</comments><slash:comments>0</slash:comments><wfw:commentRss>http://www.blogjava.net/fullfocus/comments/commentRss/199608.html</wfw:commentRss><trackback:ping>http://www.blogjava.net/fullfocus/services/trackbacks/199608.html</trackback:ping><description><![CDATA[&nbsp;&nbsp;&nbsp;&nbsp; 摘要: 从现在开始，开始我真正的研究计划：聚类获取有价值的信息。<br>今天开始详细了解了已有的资源和项目代码，包含中文分词，文档向量化，现有的KA+K-MEANS聚类算法。但是据观察，聚类效果上不是很满意，主要是类簇的关键字抽取不够准确，特征选择尚未考虑，聚类精度需要提高。<br>以下是现有系统的流程图：明天开始到下周末，研究<br>1.ka+k-means，其他k-means方法，找出系统不足点<br>2.研究特征选择方法，提高聚类前数据的质量&nbsp;&nbsp;<a href='http://www.blogjava.net/fullfocus/archive/2008/05/09/199608.html'>阅读全文</a><img src ="http://www.blogjava.net/fullfocus/aggbug/199608.html" width = "1" height = "1" /><br><br><div align=right><a style="text-decoration:none;" href="http://www.blogjava.net/fullfocus/" target="_blank">fullfocus</a> 2008-05-09 22:22 <a href="http://www.blogjava.net/fullfocus/archive/2008/05/09/199608.html#Feedback" target="_blank" style="text-decoration:none;">发表评论</a></div>]]></description></item><item><title>K-MEANS,AHC, single path直观演示---Clustering Web Search Results</title><link>http://www.blogjava.net/fullfocus/archive/2008/05/08/199259.html</link><dc:creator>fullfocus</dc:creator><author>fullfocus</author><pubDate>Thu, 08 May 2008 07:21:00 GMT</pubDate><guid>http://www.blogjava.net/fullfocus/archive/2008/05/08/199259.html</guid><wfw:comment>http://www.blogjava.net/fullfocus/comments/199259.html</wfw:comment><comments>http://www.blogjava.net/fullfocus/archive/2008/05/08/199259.html#Feedback</comments><slash:comments>0</slash:comments><wfw:commentRss>http://www.blogjava.net/fullfocus/comments/commentRss/199259.html</wfw:commentRss><trackback:ping>http://www.blogjava.net/fullfocus/services/trackbacks/199259.html</trackback:ping><description><![CDATA[<br />
<iframe src='http://docs.google.com/EmbedSlideshow?docid=dfrm2nb7_107c6xzpjfd' frameborder='0' width='410' height='342'></iframe>


<img src ="http://www.blogjava.net/fullfocus/aggbug/199259.html" width = "1" height = "1" /><br><br><div align=right><a style="text-decoration:none;" href="http://www.blogjava.net/fullfocus/" target="_blank">fullfocus</a> 2008-05-08 15:21 <a href="http://www.blogjava.net/fullfocus/archive/2008/05/08/199259.html#Feedback" target="_blank" style="text-decoration:none;">发表评论</a></div>]]></description></item><item><title>TF-IDF解释</title><link>http://www.blogjava.net/fullfocus/archive/2008/05/07/198987.html</link><dc:creator>fullfocus</dc:creator><author>fullfocus</author><pubDate>Wed, 07 May 2008 07:38:00 GMT</pubDate><guid>http://www.blogjava.net/fullfocus/archive/2008/05/07/198987.html</guid><wfw:comment>http://www.blogjava.net/fullfocus/comments/198987.html</wfw:comment><comments>http://www.blogjava.net/fullfocus/archive/2008/05/07/198987.html#Feedback</comments><slash:comments>0</slash:comments><wfw:commentRss>http://www.blogjava.net/fullfocus/comments/commentRss/198987.html</wfw:commentRss><trackback:ping>http://www.blogjava.net/fullfocus/services/trackbacks/198987.html</trackback:ping><description><![CDATA[&nbsp;&nbsp;&nbsp;&nbsp; 摘要: google的数学之美 系列九 -- 如何确定网页和查询的相关性<br>主要讲解TF-IDF技术，与判断查询相关性。&nbsp;&nbsp;<a href='http://www.blogjava.net/fullfocus/archive/2008/05/07/198987.html'>阅读全文</a><img src ="http://www.blogjava.net/fullfocus/aggbug/198987.html" width = "1" height = "1" /><br><br><div align=right><a style="text-decoration:none;" href="http://www.blogjava.net/fullfocus/" target="_blank">fullfocus</a> 2008-05-07 15:38 <a href="http://www.blogjava.net/fullfocus/archive/2008/05/07/198987.html#Feedback" target="_blank" style="text-decoration:none;">发表评论</a></div>]]></description></item><item><title>召回率与精度</title><link>http://www.blogjava.net/fullfocus/archive/2008/05/07/198963.html</link><dc:creator>fullfocus</dc:creator><author>fullfocus</author><pubDate>Wed, 07 May 2008 06:35:00 GMT</pubDate><guid>http://www.blogjava.net/fullfocus/archive/2008/05/07/198963.html</guid><wfw:comment>http://www.blogjava.net/fullfocus/comments/198963.html</wfw:comment><comments>http://www.blogjava.net/fullfocus/archive/2008/05/07/198963.html#Feedback</comments><slash:comments>1</slash:comments><wfw:commentRss>http://www.blogjava.net/fullfocus/comments/commentRss/198963.html</wfw:commentRss><trackback:ping>http://www.blogjava.net/fullfocus/services/trackbacks/198963.html</trackback:ping><description><![CDATA[&nbsp;&nbsp;&nbsp;&nbsp; 摘要: 一直搞不清搜索引擎的查全率和查准率是什么意思,只知道这两个是衡量一个搜索引擎性能的. 今个 看一篇 南大的学士论文的时候, 又碰到这个问题. 所以决定把他搞清楚, 上百度搜了一下, 所获很多. &nbsp;&nbsp;<a href='http://www.blogjava.net/fullfocus/archive/2008/05/07/198963.html'>阅读全文</a><img src ="http://www.blogjava.net/fullfocus/aggbug/198963.html" width = "1" height = "1" /><br><br><div align=right><a style="text-decoration:none;" href="http://www.blogjava.net/fullfocus/" target="_blank">fullfocus</a> 2008-05-07 14:35 <a href="http://www.blogjava.net/fullfocus/archive/2008/05/07/198963.html#Feedback" target="_blank" style="text-decoration:none;">发表评论</a></div>]]></description></item></channel></rss>